3 SPDX-License-Identifier: BSD-2-Clause
5 Copyright (c) 2018-2021 Gavin D. Howard and contributors.
7 Redistribution and use in source and binary forms, with or without
8 modification, are permitted provided that the following conditions are met:
10 * Redistributions of source code must retain the above copyright notice, this
11 list of conditions and the following disclaimer.
13 * Redistributions in binary form must reproduce the above copyright notice,
14 this list of conditions and the following disclaimer in the documentation
15 and/or other materials provided with the distribution.
17 THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
18 AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
19 IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
20 ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE
21 LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
22 CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
23 SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
24 INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
25 CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
26 ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
27 POSSIBILITY OF SUCH DAMAGE.
33 bc - arbitrary-precision decimal arithmetic language and calculator
37 **bc** [**-ghilPqRsvVw**] [**-\-global-stacks**] [**-\-help**] [**-\-interactive**] [**-\-mathlib**] [**-\-no-prompt**] [**-\-no-read-prompt**] [**-\-quiet**] [**-\-standard**] [**-\-warn**] [**-\-version**] [**-e** *expr*] [**-\-expression**=*expr*...] [**-f** *file*...] [**-\-file**=*file*...] [*file*...]
41 bc(1) is an interactive processor for a language first standardized in 1991 by
42 POSIX. (The current standard is [here][1].) The language provides unlimited
43 precision decimal arithmetic and is somewhat C-like, but there are differences.
44 Such differences will be noted in this document.
46 After parsing and handling options, this bc(1) reads any files given on the
47 command line and executes them before reading from **stdin**.
51 The following are the options that bc(1) accepts.
53 **-g**, **-\-global-stacks**
55 Turns the globals **ibase**, **obase**, and **scale** into stacks.
57 This has the effect that a copy of the current value of all three are pushed
58 onto a stack for every function call, as well as popped when every function
59 returns. This means that functions can assign to any and all of those
60 globals without worrying that the change will affect other functions.
61 Thus, a hypothetical function named **output(x,b)** that simply printed
62 **x** in base **b** could be written like this:
64 define void output(x, b) {
71 define void output(x, b) {
79 This makes writing functions much easier.
81 However, since using this flag means that functions cannot set **ibase**,
82 **obase**, or **scale** globally, functions that are made to do so cannot
83 work anymore. There are two possible use cases for that, and each has a
86 First, if a function is called on startup to turn bc(1) into a number
87 converter, it is possible to replace that capability with various shell
90 alias d2o="bc -e ibase=A -e obase=8"
91 alias h2b="bc -e ibase=G -e obase=2"
93 Second, if the purpose of a function is to set **ibase**, **obase**, or
94 **scale** globally for any other purpose, it could be split into one to
95 three functions (based on how many globals it sets) and each of those
96 functions could return the desired value for a global.
98 If the behavior of this option is desired for every run of bc(1), then users
99 could make sure to define **BC_ENV_ARGS** and include this option (see the
100 **ENVIRONMENT VARIABLES** section for more details).
102 If **-s**, **-w**, or any equivalents are used, this option is ignored.
104 This is a **non-portable extension**.
108 : Prints a usage message and quits.
110 **-i**, **-\-interactive**
112 : Forces interactive mode. (See the **INTERACTIVE MODE** section.)
114 This is a **non-portable extension**.
116 **-l**, **-\-mathlib**
118 : Sets **scale** (see the **SYNTAX** section) to **20** and loads the included
119 math library before running any code, including any expressions or files
120 specified on the command line.
122 To learn what is in the library, see the **LIBRARY** section.
124 **-P**, **-\-no-prompt**
126 : This option is a no-op.
128 This is a **non-portable extension**.
130 **-R**, **-\-no-read-prompt**
132 : Because bc(1) was built without support for prompts, this option is a no-op.
134 This is a **non-portable extension**.
138 : This option is for compatibility with the [GNU bc(1)][2]; it is a no-op.
139 Without this option, GNU bc(1) prints a copyright header. This bc(1) only
140 prints the copyright header if one or more of the **-v**, **-V**, or
141 **-\-version** options are given.
143 This is a **non-portable extension**.
145 **-s**, **-\-standard**
147 : Process exactly the language defined by the [standard][1] and error if any
150 This is a **non-portable extension**.
152 **-v**, **-V**, **-\-version**
154 : Print the version information (copyright header) and exit.
156 This is a **non-portable extension**.
160 : Like **-s** and **-\-standard**, except that warnings (and not errors) are
161 printed for non-standard extensions and execution continues normally.
163 This is a **non-portable extension**.
165 **-e** *expr*, **-\-expression**=*expr*
167 : Evaluates *expr*. If multiple expressions are given, they are evaluated in
168 order. If files are given as well (see below), the expressions and files are
169 evaluated in the order given. This means that if a file is given before an
170 expression, the file is read in and evaluated first.
172 If this option is given on the command-line (i.e., not in **BC_ENV_ARGS**,
173 see the **ENVIRONMENT VARIABLES** section), then after processing all
174 expressions and files, bc(1) will exit, unless **-** (**stdin**) was given
175 as an argument at least once to **-f** or **-\-file**, whether on the
176 command-line or in **BC_ENV_ARGS**. However, if any other **-e**,
177 **-\-expression**, **-f**, or **-\-file** arguments are given after **-f-**
178 or equivalent is given, bc(1) will give a fatal error and exit.
180 This is a **non-portable extension**.
182 **-f** *file*, **-\-file**=*file*
184 : Reads in *file* and evaluates it, line by line, as though it were read
185 through **stdin**. If expressions are also given (see above), the
186 expressions are evaluated in the order given.
188 If this option is given on the command-line (i.e., not in **BC_ENV_ARGS**,
189 see the **ENVIRONMENT VARIABLES** section), then after processing all
190 expressions and files, bc(1) will exit, unless **-** (**stdin**) was given
191 as an argument at least once to **-f** or **-\-file**. However, if any other
192 **-e**, **-\-expression**, **-f**, or **-\-file** arguments are given after
193 **-f-** or equivalent is given, bc(1) will give a fatal error and exit.
195 This is a **non-portable extension**.
197 All long options are **non-portable extensions**.
201 Any non-error output is written to **stdout**. In addition, if history (see the
202 **HISTORY** section) and the prompt (see the **TTY MODE** section) are enabled,
203 both are output to **stdout**.
205 **Note**: Unlike other bc(1) implementations, this bc(1) will issue a fatal
206 error (see the **EXIT STATUS** section) if it cannot write to **stdout**, so if
207 **stdout** is closed, as in **bc <file> >&-**, it will quit with an error. This
208 is done so that bc(1) can report problems when **stdout** is redirected to a
211 If there are scripts that depend on the behavior of other bc(1) implementations,
212 it is recommended that those scripts be changed to redirect **stdout** to
217 Any error output is written to **stderr**.
219 **Note**: Unlike other bc(1) implementations, this bc(1) will issue a fatal
220 error (see the **EXIT STATUS** section) if it cannot write to **stderr**, so if
221 **stderr** is closed, as in **bc <file> 2>&-**, it will quit with an error. This
222 is done so that bc(1) can exit with an error code when **stderr** is redirected
225 If there are scripts that depend on the behavior of other bc(1) implementations,
226 it is recommended that those scripts be changed to redirect **stderr** to
231 The syntax for bc(1) programs is mostly C-like, with some differences. This
232 bc(1) follows the [POSIX standard][1], which is a much more thorough resource
233 for the language this bc(1) accepts. This section is meant to be a summary and a
234 listing of all the extensions to the standard.
236 In the sections below, **E** means expression, **S** means statement, and **I**
239 Identifiers (**I**) start with a lowercase letter and can be followed by any
240 number (up to **BC_NAME_MAX-1**) of lowercase letters (**a-z**), digits
241 (**0-9**), and underscores (**\_**). The regex is **\[a-z\]\[a-z0-9\_\]\***.
242 Identifiers with more than one character (letter) are a
243 **non-portable extension**.
245 **ibase** is a global variable determining how to interpret constant numbers. It
246 is the "input" base, or the number base used for interpreting input numbers.
247 **ibase** is initially **10**. If the **-s** (**-\-standard**) and **-w**
248 (**-\-warn**) flags were not given on the command line, the max allowable value
249 for **ibase** is **36**. Otherwise, it is **16**. The min allowable value for
250 **ibase** is **2**. The max allowable value for **ibase** can be queried in
251 bc(1) programs with the **maxibase()** built-in function.
253 **obase** is a global variable determining how to output results. It is the
254 "output" base, or the number base used for outputting numbers. **obase** is
255 initially **10**. The max allowable value for **obase** is **BC_BASE_MAX** and
256 can be queried in bc(1) programs with the **maxobase()** built-in function. The
257 min allowable value for **obase** is **2**. Values are output in the specified
260 The *scale* of an expression is the number of digits in the result of the
261 expression right of the decimal point, and **scale** is a global variable that
262 sets the precision of any operations, with exceptions. **scale** is initially
263 **0**. **scale** cannot be negative. The max allowable value for **scale** is
264 **BC_SCALE_MAX** and can be queried in bc(1) programs with the **maxscale()**
267 bc(1) has both *global* variables and *local* variables. All *local*
268 variables are local to the function; they are parameters or are introduced in
269 the **auto** list of a function (see the **FUNCTIONS** section). If a variable
270 is accessed which is not a parameter or in the **auto** list, it is assumed to
271 be *global*. If a parent function has a *local* variable version of a variable
272 that a child function considers *global*, the value of that *global* variable in
273 the child function is the value of the variable in the parent function, not the
274 value of the actual *global* variable.
276 All of the above applies to arrays as well.
278 The value of a statement that is an expression (i.e., any of the named
279 expressions or operands) is printed unless the lowest precedence operator is an
280 assignment operator *and* the expression is notsurrounded by parentheses.
282 The value that is printed is also assigned to the special variable **last**. A
283 single dot (**.**) may also be used as a synonym for **last**. These are
284 **non-portable extensions**.
286 Either semicolons or newlines may separate statements.
290 There are two kinds of comments:
292 1. Block comments are enclosed in **/\*** and **\*/**.
293 2. Line comments go from **#** until, and not including, the next newline. This
294 is a **non-portable extension**.
298 The following are named expressions in bc(1):
301 2. Array Elements: **I[E]**
305 6. **last** or a single dot (**.**)
307 Number 6 is a **non-portable extension**.
309 Variables and arrays do not interfere; users can have arrays named the same as
310 variables. This also applies to functions (see the **FUNCTIONS** section), so a
311 user can have a variable, array, and function that all have the same name, and
312 they will not shadow each other, whether inside of functions or not.
314 Named expressions are required as the operand of **increment**/**decrement**
315 operators and as the left side of **assignment** operators (see the *Operators*
320 The following are valid operands in bc(1):
322 1. Numbers (see the *Numbers* subsection below).
323 2. Array indices (**I[E]**).
324 3. **(E)**: The value of **E** (used to change precedence).
325 4. **sqrt(E)**: The square root of **E**. **E** must be non-negative.
326 5. **length(E)**: The number of significant decimal digits in **E**.
327 6. **length(I[])**: The number of elements in the array **I**. This is a
328 **non-portable extension**.
329 7. **scale(E)**: The *scale* of **E**.
330 8. **abs(E)**: The absolute value of **E**. This is a **non-portable
332 9. **I()**, **I(E)**, **I(E, E)**, and so on, where **I** is an identifier for
333 a non-**void** function (see the *Void Functions* subsection of the
334 **FUNCTIONS** section). The **E** argument(s) may also be arrays of the form
335 **I[]**, which will automatically be turned into array references (see the
336 *Array References* subsection of the **FUNCTIONS** section) if the
337 corresponding parameter in the function definition is an array reference.
338 10. **read()**: Reads a line from **stdin** and uses that as an expression. The
339 result of that expression is the result of the **read()** operand. This is a
340 **non-portable extension**.
341 11. **maxibase()**: The max allowable **ibase**. This is a **non-portable
343 12. **maxobase()**: The max allowable **obase**. This is a **non-portable
345 13. **maxscale()**: The max allowable **scale**. This is a **non-portable
350 Numbers are strings made up of digits, uppercase letters, and at most **1**
351 period for a radix. Numbers can have up to **BC_NUM_MAX** digits. Uppercase
352 letters are equal to **9** + their position in the alphabet (i.e., **A** equals
353 **10**, or **9+1**). If a digit or letter makes no sense with the current value
354 of **ibase**, they are set to the value of the highest valid digit in **ibase**.
356 Single-character numbers (i.e., **A** alone) take the value that they would have
357 if they were valid digits, regardless of the value of **ibase**. This means that
358 **A** alone always equals decimal **10** and **Z** alone always equals decimal
363 The following arithmetic and logical operators can be used. They are listed in
364 order of decreasing precedence. Operators in the same group have the same
369 : Type: Prefix and Postfix
373 Description: **increment**, **decrement**
381 Description: **negation**, **boolean not**
389 Description: **power**
397 Description: **multiply**, **divide**, **modulus**
405 Description: **add**, **subtract**
407 **=** **+=** **-=** **\*=** **/=** **%=** **\^=**
413 Description: **assignment**
415 **==** **\<=** **\>=** **!=** **\<** **\>**
421 Description: **relational**
429 Description: **boolean and**
437 Description: **boolean or**
439 The operators will be described in more detail below.
443 : The prefix and postfix **increment** and **decrement** operators behave
444 exactly like they would in C. They require a named expression (see the
445 *Named Expressions* subsection) as an operand.
447 The prefix versions of these operators are more efficient; use them where
452 : The **negation** operator returns **0** if a user attempts to negate any
453 expression with the value **0**. Otherwise, a copy of the expression with
454 its sign flipped is returned.
458 : The **boolean not** operator returns **1** if the expression is **0**, or
461 This is a **non-portable extension**.
465 : The **power** operator (not the **exclusive or** operator, as it would be in
466 C) takes two expressions and raises the first to the power of the value of
467 the second. The *scale* of the result is equal to **scale**.
469 The second expression must be an integer (no *scale*), and if it is
470 negative, the first value must be non-zero.
474 : The **multiply** operator takes two expressions, multiplies them, and
475 returns the product. If **a** is the *scale* of the first expression and
476 **b** is the *scale* of the second expression, the *scale* of the result is
477 equal to **min(a+b,max(scale,a,b))** where **min()** and **max()** return
482 : The **divide** operator takes two expressions, divides them, and returns the
483 quotient. The *scale* of the result shall be the value of **scale**.
485 The second expression must be non-zero.
489 : The **modulus** operator takes two expressions, **a** and **b**, and
490 evaluates them by 1) Computing **a/b** to current **scale** and 2) Using the
491 result of step 1 to calculate **a-(a/b)\*b** to *scale*
492 **max(scale+scale(b),scale(a))**.
494 The second expression must be non-zero.
498 : The **add** operator takes two expressions, **a** and **b**, and returns the
499 sum, with a *scale* equal to the max of the *scale*s of **a** and **b**.
503 : The **subtract** operator takes two expressions, **a** and **b**, and
504 returns the difference, with a *scale* equal to the max of the *scale*s of
507 **=** **+=** **-=** **\*=** **/=** **%=** **\^=**
509 : The **assignment** operators take two expressions, **a** and **b** where
510 **a** is a named expression (see the *Named Expressions* subsection).
512 For **=**, **b** is copied and the result is assigned to **a**. For all
513 others, **a** and **b** are applied as operands to the corresponding
514 arithmetic operator and the result is assigned to **a**.
516 **==** **\<=** **\>=** **!=** **\<** **\>**
518 : The **relational** operators compare two expressions, **a** and **b**, and
519 if the relation holds, according to C language semantics, the result is
520 **1**. Otherwise, it is **0**.
522 Note that unlike in C, these operators have a lower precedence than the
523 **assignment** operators, which means that **a=b\>c** is interpreted as
526 Also, unlike the [standard][1] requires, these operators can appear anywhere
527 any other expressions can be used. This allowance is a
528 **non-portable extension**.
532 : The **boolean and** operator takes two expressions and returns **1** if both
533 expressions are non-zero, **0** otherwise.
535 This is *not* a short-circuit operator.
537 This is a **non-portable extension**.
541 : The **boolean or** operator takes two expressions and returns **1** if one
542 of the expressions is non-zero, **0** otherwise.
544 This is *not* a short-circuit operator.
546 This is a **non-portable extension**.
550 The following items are statements:
553 2. **{** **S** **;** ... **;** **S** **}**
554 3. **if** **(** **E** **)** **S**
555 4. **if** **(** **E** **)** **S** **else** **S**
556 5. **while** **(** **E** **)** **S**
557 6. **for** **(** **E** **;** **E** **;** **E** **)** **S**
558 7. An empty statement
564 13. A string of characters, enclosed in double quotes
565 14. **print** **E** **,** ... **,** **E**
566 15. **I()**, **I(E)**, **I(E, E)**, and so on, where **I** is an identifier for
567 a **void** function (see the *Void Functions* subsection of the
568 **FUNCTIONS** section). The **E** argument(s) may also be arrays of the form
569 **I[]**, which will automatically be turned into array references (see the
570 *Array References* subsection of the **FUNCTIONS** section) if the
571 corresponding parameter in the function definition is an array reference.
573 Numbers 4, 9, 11, 12, 14, and 15 are **non-portable extensions**.
575 Also, as a **non-portable extension**, any or all of the expressions in the
576 header of a for loop may be omitted. If the condition (second expression) is
577 omitted, it is assumed to be a constant **1**.
579 The **break** statement causes a loop to stop iterating and resume execution
580 immediately following a loop. This is only allowed in loops.
582 The **continue** statement causes a loop iteration to stop early and returns to
583 the start of the loop, including testing the loop condition. This is only
586 The **if** **else** statement does the same thing as in C.
588 The **quit** statement causes bc(1) to quit, even if it is on a branch that will
589 not be executed (it is a compile-time command).
591 The **halt** statement causes bc(1) to quit, if it is executed. (Unlike **quit**
592 if it is on a branch of an **if** statement that is not executed, bc(1) does not
595 The **limits** statement prints the limits that this bc(1) is subject to. This
596 is like the **quit** statement in that it is a compile-time command.
598 An expression by itself is evaluated and printed, followed by a newline.
602 The "expressions" in a **print** statement may also be strings. If they are, there
603 are backslash escape sequences that are interpreted specially. What those
604 sequences are, and what they cause to be printed, are shown below:
618 Any other character following a backslash causes the backslash and character to
621 Any non-string expression in a print statement shall be assigned to **last**,
622 like any other expression that is printed.
624 ## Order of Evaluation
626 All expressions in a statment are evaluated left to right, except as necessary
627 to maintain order of operations. This means, for example, assuming that **i** is
628 equal to **0**, in the expression
632 the first (or 0th) element of **a** is set to **1**, and **i** is equal to **2**
633 at the end of the expression.
635 This includes function arguments. Thus, assuming **i** is equal to **0**, this
636 means that in the expression
640 the first argument passed to **x()** is **0**, and the second argument is **1**,
641 while **i** is equal to **2** before the function starts executing.
645 Function definitions are as follows:
655 Any **I** in the parameter list or **auto** list may be replaced with **I[]** to
656 make a parameter or **auto** var an array, and any **I** in the parameter list
657 may be replaced with **\*I[]** to make a parameter an array reference. Callers
658 of functions that take array references should not put an asterisk in the call;
659 they must be called with just **I[]** like normal array parameters and will be
660 automatically converted into references.
662 As a **non-portable extension**, the opening brace of a **define** statement may
663 appear on the next line.
665 As a **non-portable extension**, the return statement may also be in one of the
669 2. **return** **(** **)**
672 The first two, or not specifying a **return** statement, is equivalent to
673 **return (0)**, unless the function is a **void** function (see the *Void
674 Functions* subsection below).
678 Functions can also be **void** functions, defined as follows:
681 define void I(I,...,I){
688 They can only be used as standalone expressions, where such an expression would
689 be printed alone, except in a print statement.
691 Void functions can only use the first two **return** statements listed above.
692 They can also omit the return statement entirely.
694 The word "void" is not treated as a keyword; it is still possible to have
695 variables, arrays, and functions named **void**. The word "void" is only
696 treated specially right after the **define** keyword.
698 This is a **non-portable extension**.
702 For any array in the parameter list, if the array is declared in the form
708 it is a **reference**. Any changes to the array in the function are reflected,
709 when the function returns, to the array that was passed in.
711 Other than this, all function arguments are passed by value.
713 This is a **non-portable extension**.
717 All of the functions below are available when the **-l** or **-\-mathlib**
718 command-line flags are given.
722 The [standard][1] defines the following functions for the math library:
726 : Returns the sine of **x**, which is assumed to be in radians.
728 This is a transcendental function (see the *Transcendental Functions*
733 : Returns the cosine of **x**, which is assumed to be in radians.
735 This is a transcendental function (see the *Transcendental Functions*
740 : Returns the arctangent of **x**, in radians.
742 This is a transcendental function (see the *Transcendental Functions*
747 : Returns the natural logarithm of **x**.
749 This is a transcendental function (see the *Transcendental Functions*
754 : Returns the mathematical constant **e** raised to the power of **x**.
756 This is a transcendental function (see the *Transcendental Functions*
761 : Returns the bessel integer order **n** (truncated) of **x**.
763 This is a transcendental function (see the *Transcendental Functions*
766 ## Transcendental Functions
768 All transcendental functions can return slightly inaccurate results (up to 1
769 [ULP][4]). This is unavoidable, and [this article][5] explains why it is
770 impossible and unnecessary to calculate exact results for the transcendental
773 Because of the possible inaccuracy, I recommend that users call those functions
774 with the precision (**scale**) set to at least 1 higher than is necessary. If
775 exact results are *absolutely* required, users can double the precision
776 (**scale**) and then truncate.
778 The transcendental functions in the standard math library are:
789 When bc(1) encounters an error or a signal that it has a non-default handler
790 for, it resets. This means that several things happen.
792 First, any functions that are executing are stopped and popped off the stack.
793 The behavior is not unlike that of exceptions in programming languages. Then
794 the execution point is set so that any code waiting to execute (after all
795 functions returned) is skipped.
797 Thus, when bc(1) resets, it skips any remaining code waiting to be executed.
798 Then, if it is interactive mode, and the error was not a fatal error (see the
799 **EXIT STATUS** section), it asks for more input; otherwise, it exits with the
800 appropriate return code.
802 Note that this reset behavior is different from the GNU bc(1), which attempts to
803 start executing the statement right after the one that caused an error.
807 Most bc(1) implementations use **char** types to calculate the value of **1**
808 decimal digit at a time, but that can be slow. This bc(1) does something
811 It uses large integers to calculate more than **1** decimal digit at a time. If
812 built in a environment where **BC_LONG_BIT** (see the **LIMITS** section) is
813 **64**, then each integer has **9** decimal digits. If built in an environment
814 where **BC_LONG_BIT** is **32** then each integer has **4** decimal digits. This
815 value (the number of decimal digits per large integer) is called
818 The actual values of **BC_LONG_BIT** and **BC_BASE_DIGS** can be queried with
819 the **limits** statement.
821 In addition, this bc(1) uses an even larger integer for overflow checking. This
822 integer type depends on the value of **BC_LONG_BIT**, but is always at least
823 twice as large as the integer type used to store digits.
827 The following are the limits on bc(1):
831 : The number of bits in the **long** type in the environment where bc(1) was
832 built. This determines how many decimal digits can be stored in a single
833 large integer (see the **PERFORMANCE** section).
837 : The number of decimal digits per large integer (see the **PERFORMANCE**
838 section). Depends on **BC_LONG_BIT**.
842 : The max decimal number that each large integer can store (see
843 **BC_BASE_DIGS**) plus **1**. Depends on **BC_BASE_DIGS**.
847 : The max number that the overflow type (see the **PERFORMANCE** section) can
848 hold. Depends on **BC_LONG_BIT**.
852 : The maximum output base. Set at **BC_BASE_POW**.
856 : The maximum size of arrays. Set at **SIZE_MAX-1**.
860 : The maximum **scale**. Set at **BC_OVERFLOW_MAX-1**.
864 : The maximum length of strings. Set at **BC_OVERFLOW_MAX-1**.
868 : The maximum length of identifiers. Set at **BC_OVERFLOW_MAX-1**.
872 : The maximum length of a number (in decimal digits), which includes digits
873 after the decimal point. Set at **BC_OVERFLOW_MAX-1**.
877 : The maximum allowable exponent (positive or negative). Set at
882 : The maximum number of vars/arrays. Set at **SIZE_MAX-1**.
884 The actual values can be queried with the **limits** statement.
886 These limits are meant to be effectively non-existent; the limits are so large
887 (at least on 64-bit machines) that there should not be any point at which they
888 become a problem. In fact, memory should be exhausted before these limits should
891 # ENVIRONMENT VARIABLES
893 bc(1) recognizes the following environment variables:
897 : If this variable exists (no matter the contents), bc(1) behaves as if
898 the **-s** option was given.
902 : This is another way to give command-line arguments to bc(1). They should be
903 in the same format as all other command-line arguments. These are always
904 processed first, so any files given in **BC_ENV_ARGS** will be processed
905 before arguments and files given on the command-line. This gives the user
906 the ability to set up "standard" options and files to be used at every
907 invocation. The most useful thing for such files to contain would be useful
908 functions that the user might want every time bc(1) runs.
910 The code that parses **BC_ENV_ARGS** will correctly handle quoted arguments,
911 but it does not understand escape sequences. For example, the string
912 **"/home/gavin/some bc file.bc"** will be correctly parsed, but the string
913 **"/home/gavin/some \"bc\" file.bc"** will include the backslashes.
915 The quote parsing will handle either kind of quotes, **'** or **"**. Thus,
916 if you have a file with any number of single quotes in the name, you can use
917 double quotes as the outside quotes, as in **"some 'bc' file.bc"**, and vice
918 versa if you have a file with double quotes. However, handling a file with
919 both kinds of quotes in **BC_ENV_ARGS** is not supported due to the
920 complexity of the parsing, though such files are still supported on the
921 command-line where the parsing is done by the shell.
925 : If this environment variable exists and contains an integer that is greater
926 than **1** and is less than **UINT16_MAX** (**2\^16-1**), bc(1) will output
927 lines to that length, including the backslash (**\\**). The default line
932 bc(1) returns the following exit statuses:
940 : A math error occurred. This follows standard practice of using **1** for
941 expected errors, since math errors will happen in the process of normal
944 Math errors include divide by **0**, taking the square root of a negative
945 number, attempting to convert a negative number to a hardware integer,
946 overflow when converting a number to a hardware integer, and attempting to
947 use a non-integer where an integer is required.
949 Converting to a hardware integer happens for the second operand of the power
950 (**\^**) operator and the corresponding assignment operator.
954 : A parse error occurred.
956 Parse errors include unexpected **EOF**, using an invalid character, failing
957 to find the end of a string or comment, using a token where it is invalid,
958 giving an invalid expression, giving an invalid print statement, giving an
959 invalid function definition, attempting to assign to an expression that is
960 not a named expression (see the *Named Expressions* subsection of the
961 **SYNTAX** section), giving an invalid **auto** list, having a duplicate
962 **auto**/function parameter, failing to find the end of a code block,
963 attempting to return a value from a **void** function, attempting to use a
964 variable as a reference, and using any extensions when the option **-s** or
965 any equivalents were given.
969 : A runtime error occurred.
971 Runtime errors include assigning an invalid number to **ibase**, **obase**,
972 or **scale**; give a bad expression to a **read()** call, calling **read()**
973 inside of a **read()** call, type errors, passing the wrong number of
974 arguments to functions, attempting to call an undefined function, and
975 attempting to use a **void** function call as a value in an expression.
979 : A fatal error occurred.
981 Fatal errors include memory allocation errors, I/O errors, failing to open
982 files, attempting to use files that do not have only ASCII characters (bc(1)
983 only accepts ASCII characters), attempting to open a directory as a file,
984 and giving invalid command-line options.
986 The exit status **4** is special; when a fatal error occurs, bc(1) always exits
987 and returns **4**, no matter what mode bc(1) is in.
989 The other statuses will only be returned when bc(1) is not in interactive mode
990 (see the **INTERACTIVE MODE** section), since bc(1) resets its state (see the
991 **RESET** section) and accepts more input when one of those errors occurs in
992 interactive mode. This is also the case when interactive mode is forced by the
993 **-i** flag or **-\-interactive** option.
995 These exit statuses allow bc(1) to be used in shell scripting with error
996 checking, and its normal behavior can be forced by using the **-i** flag or
997 **-\-interactive** option.
1001 Per the [standard][1], bc(1) has an interactive mode and a non-interactive mode.
1002 Interactive mode is turned on automatically when both **stdin** and **stdout**
1003 are hooked to a terminal, but the **-i** flag and **-\-interactive** option can
1004 turn it on in other cases.
1006 In interactive mode, bc(1) attempts to recover from errors (see the **RESET**
1007 section), and in normal execution, flushes **stdout** as soon as execution is
1008 done for the current input.
1012 If **stdin**, **stdout**, and **stderr** are all connected to a TTY, bc(1) turns
1015 TTY mode is different from interactive mode because interactive mode is required
1016 in the [bc(1) specification][1], and interactive mode requires only **stdin**
1017 and **stdout** to be connected to a terminal.
1021 Sending a **SIGINT** will cause bc(1) to stop execution of the current input. If
1022 bc(1) is in TTY mode (see the **TTY MODE** section), it will reset (see the
1023 **RESET** section). Otherwise, it will clean up and exit.
1025 Note that "current input" can mean one of two things. If bc(1) is processing
1026 input from **stdin** in TTY mode, it will ask for more input. If bc(1) is
1027 processing input from a file in TTY mode, it will stop processing the file and
1028 start processing the next file, if one exists, or ask for input from **stdin**
1029 if no other file exists.
1031 This means that if a **SIGINT** is sent to bc(1) as it is executing a file, it
1032 can seem as though bc(1) did not respond to the signal since it will immediately
1033 start executing the next file. This is by design; most files that users execute
1034 when interacting with bc(1) have function definitions, which are quick to parse.
1035 If a file takes a long time to execute, there may be a bug in that file. The
1036 rest of the files could still be executed without problem, allowing the user to
1039 **SIGTERM** and **SIGQUIT** cause bc(1) to clean up and exit, and it uses the
1040 default handler for all other signals.
1044 This bc(1) ships with support for adding error messages for different locales
1045 and thus, supports **LC_MESSAGES**.
1053 bc(1) is compliant with the [IEEE Std 1003.1-2017 (“POSIX.1-2017”)][1]
1054 specification. The flags **-efghiqsvVw**, all long options, and the extensions
1055 noted above are extensions to that specification.
1057 Note that the specification explicitly says that bc(1) only accepts numbers that
1058 use a period (**.**) as a radix point, regardless of the value of
1061 This bc(1) supports error messages for different locales, and thus, it supports
1066 None are known. Report bugs at https://git.yzena.com/gavin/bc.
1070 Gavin D. Howard <gavin@yzena.com> and contributors.
1072 [1]: https://pubs.opengroup.org/onlinepubs/9699919799/utilities/bc.html
1073 [2]: https://www.gnu.org/software/bc/
1074 [3]: https://en.wikipedia.org/wiki/Rounding#Round_half_away_from_zero
1075 [4]: https://en.wikipedia.org/wiki/Unit_in_the_last_place
1076 [5]: https://people.eecs.berkeley.edu/~wkahan/LOG10HAF.TXT
1077 [6]: https://en.wikipedia.org/wiki/Rounding#Rounding_away_from_zero